中文摘要：

随着互联网上三维模型的大量产生,从已有的模型中找到目标模型,就成为一个迫切的实际问题.根据三维模型投影轮廓图像的特点,提出了边界点描述符特征.在三维模型规范化和正交投影的基础上,采用边界点特征进行三维模型检索.利用普林斯顿大学三维模型库的十类模型作为实验数据,与射线法、D2形分布方法、Zernike矩方法进行对比实验,结果表明：该种特征稳定,在人造物类别和植物类别上提高了检索的准确性.

英文摘要：

With the proliferation of 3D models on the Internet, it becomes an emergency task to obtain the desired models from the existing. This paper proposed a kind of boundary descriptor according to the characteristics of the contour images of projected 3D models. This descriptor is applied in 3D model retrieval by combining with normalization and orthogonal projection. Using ten categories of 3D models in PSB, and Comparing with the performance of ray-based method and D2 shape distribution method and Zernike moment method, the experiment results show that the method is stable, and improves the accuracy of retrieval in artificiality category and plant category.